Artwork for cover and divider pages created by Kimber Horne using generative A.I. in Adobe Firefly.
Find the latest vehicle production data and budget reports, industry trends, and contact information for state, national and federal agencies, manufacturers, dealers, and suppliers.
A Tennessee voter casts his ballot in Nashville during a special election this month. The U.S. Department of Justice has sent confidential draft agreements on voter data sharing to more than a dozen states, including Tennessee — part of the Trump administration’s effort to obtain unredacted voter rolls from states. (Photo by Jon Cherry/Getty Images)
The U.S. Department of Justice has sent a confidential draft agreement to more than a dozen states that would require election officials to remove any alleged ineligible voters identified during a federal review of their voter rolls.
The agreement — called a memorandum of understanding, or MOU — would hand the federal government a major role in election administration, a responsibility that belongs to the states under the U.S. Constitution.
A Justice Department official identified 11 states that have expressed an interest in the agreement during a federal court hearing in December, according to a transcript reviewed by Stateline. Two additional states, Colorado and Wisconsin, have publicly rejected the memorandum of understanding and released copies of the proposal.
The 11 states “all fall into the list of, they have expressed with us a willingness to comply based on the represented MOU that we have sent them,” Eric Neff, the acting chief of the Justice Department’s Voting Section, said at the hearing. He spoke at a Dec. 4 hearing in a federal lawsuit brought by the Justice Department against California, which has refused a demand for the state’s voter data.
Neff’s courtroom disclosure, which Stateline is the first to report, comes as the Justice Department has sued 21 states and the District of Columbia for unredacted copies of their voter rolls after demanding the data from most states in recent months. The unredacted lists include sensitive personal information, such as driver’s license and partial Social Security numbers.
The states Neff identified are led by Republicans — Alabama, Mississippi, Missouri, Montana, Nebraska, South Carolina, South Dakota, Texas, Tennessee, Utah and Virginia.
The administration’s lawsuits mostly target Democratic states, where election officials refused initial requests for voter data and allege the demand is unlawful and risks the privacy of millions of voters. They have also voiced fears that the Trump administration could use the information to target its political enemies.
Neff said four states with Republican secretaries of state — Arkansas, Indiana, Kansas and Wyoming — have “complied voluntarily” with the Justice Department’s demand without memoranda of understanding.
What the DOJ is trying to do is something that should frighten everybody across the political spectrum.
– David Becker, executive director of the Center for Election Innovation & Research
The Justice Department says it needs voters’ detailed information to ensure ineligible people are kept off state voter rolls and that only citizens are voting.
Federal officials say they will follow federal privacy laws, but critics fear voter data is being shared with the U.S. Department of Homeland Security, which operates a powerful citizenship verification tool known as SAVE. The Trump administration has previously confirmed the Justice Department plans to share voter data with Homeland Security.
“What the DOJ is trying to do is something that should frighten everybody across the political spectrum,” said David Becker, executive director of the nonpartisan Center for Election Innovation & Research. “They’re trying to use the power of the executive branch to bully states into turning over highly sensitive data: date of birth, Social Security number, driver’s license — the holy trinity of identity theft.”
Becker, who worked as a senior trial attorney in the Justice Department’s Voting Section during the Clinton and George W. Bush administrations, told reporters on Dec. 8 that several states received the memorandum. But Neff’s identification of 11 states wasn’t widely available until the judge in the California lawsuit on Tuesday ordered the transcript of the Dec. 4 hearing immediately posted to the lawsuit’s public docket, where Stateline accessed it.
A transcript of a Dec. 4 federal court hearing showing Eric Neff, acting chief of the Justice Department’s Voting Section, listing states that he says are willing to comply with a request for voter data based on a proposed memorandum of understanding (MOU). (Screenshot by Jonathan Shorman)
The draft memorandum of understanding, which is labeled “confidential,” outlines the terms of the proposed agreement between each state and the Justice Department. After a state provides its voter roll, the federal department would agree to test, analyze and assess the information. The department would then notify states of “any voter list maintenance issues, insufficiencies, inadequacies, deficiencies, anomalies, or concerns” found.
Each state would agree to “clean” its voter roll within 45 days by removing any ineligible voters, according to the memorandum. States would then resubmit their voter data to the Justice Department for verification.
While the Justice Department has demanded states’ voter rolls since this summer, the memorandum of understanding offers the most detailed picture to date of how the Trump administration plans to use the data.
“It lays out in a way that we haven’t seen in any other context their plan for one of the things, I will say, that they plan to do, which is disturbing,” said Eileen O’Connor, a senior counsel and manager in the voting rights and election program at the Brennan Center for Justice at New York University, a progressive think tank.
O’Connor was a trial attorney in the Justice Department’s Voting Section during the Obama, first Trump and Biden administrations. “I think with each passing lawsuit, they are clearly trying to create a national database of every voter in the country,” she said.
The Justice Department didn’t answer questions from Stateline about how many states had been sent the memorandum and whether any had signed it.
Assistant U.S. Attorney General Harmeet Dhillon, who leads the Justice Department’s Civil Rights Division, wrote in a statement to Stateline that the department has a statutory mandate to enforce federal voting rights laws. Ensuring the voting public’s confidence in election integrity is a top priority of the Trump administration, she wrote.
“Clean voter rolls and basic election safeguards are requisites for free, fair, and transparent elections,” Dhillon wrote.
Federal involvement in elections
The Justice Department memorandum, if implemented, would mark a significant departure from how election officials typically maintain voter rolls.
States, often in coordination with local election officials, check lists for changes in address, deaths and other reasons for ineligibility, such as a felony conviction. States typically perform this task with little to no federal involvement.
Some states participate in voluntary programs that allow election officials to share voter information with other states for the purposes of looking for voters who may have moved or who are registered in multiple locations. But those don’t include the federal government, which plays a limited role in election administration under the United States’ decentralized approach to elections.
Matt Crane, executive director of the Colorado County Clerks Association, said clerks continually look at death records and other sources of data to update voter lists. He said the United States’ localized election system is a strength that guards against election interference.
“The federal government has no role in list maintenance,” Crane said.
But that has begun to change under the Trump administration, as President Donald Trump has made removing noncitizen voters a priority.
Earlier this year, Homeland Security overhauled the SAVE program into a tool that can scan millions of voter records against government databases for evidence of citizenship. The program was previously used for one-off searches to check whether noncitizens were eligible for government benefits.
Some Republican secretaries of state have agreed to upload their voter rolls into SAVE. Democratic secretaries of state object to using the program and say they are wary of what will happen to the voter information once it’s provided to the Trump administration, including its potential use by the Department of Homeland Security.
While SAVE can flag voters with potential eligibility issues, the onus now is still on state officials to investigate whether those voters are actually ineligible and decide whether to initiate a process to remove them from the rolls.
By contrast, the Justice Department memorandum would empower federal officials to take a more active role, allowing them to check the work of state election officials as they remove — or don’t remove — voters.
“We have a system that allows Americans to voice their opinions and to hold government accountable, and that is so fundamentally central to the way our system works,” Oregon Democratic Secretary of State Tobias Read, who has been sued by the Justice Department, said in an interview. “We should be focused on how to make that better, not on erecting artificial barriers and putting people’s privacy and confidence at risk for no reason.”
Republican interest
Some GOP election officials have welcomed the Trump administration’s interest and have accused the Biden administration of not doing enough to help states vet their voter rolls. In particular, they praise the overhaul of SAVE, which some GOP secretaries of state had requested before Trump took office.
Some secretaries have touted the removal of noncitizen voters after using SAVE. Wyoming Secretary of State Chuck Gray, a Republican, in November announced three voters identified as noncitizens had been removed from his state’s voter rolls. Gray has also provided the Justice Department with full access to Wyoming’s voter roll.
“The voter list maintenance that we have been conducting is extremely important for election integrity,” Gray said in a news release.
But as of early December, nearly all states hadn’t provided the Justice Department access to their unredacted voter rolls, with Neff identifying only four that had shared their lists. It also remains unclear whether any state has signed the memorandum of understanding. No state has told Stateline it signed the document.
Nebraska Secretary of State Robert Evnen, a Republican, has received a memorandum of understanding and plans to comply with the Justice Department request, pending the outcome of an ongoing lawsuit, Evnen spokesperson Rani Taborek-Potter wrote in an email to Stateline. A voting advocacy group has sued to block the release of the data.
In an interview with Kentucky Lantern, Kentucky Republican Secretary of State Michael Adams said that his office was “going back and forth a little bit” with the Justice Department over what federal law requires.
“We’ve not really figured out exactly where that line is of what-all they’re entitled to,” Adams said. “What’s not in dispute is they’re entitled to the vast majority of information — people’s names, addresses, birthdays — and we’ve given them all of that.”
Adams added that many state officials “are in the same boat of trying to figure out what exactly they need to do their job and what our obligations are legally.”
Utah Lt. Gov. Deidre Henderson, a Republican, confirmed in a statement to Stateline that her office received a proposed memorandum of understanding from the Justice Department. “We are in the process of reviewing the document with our attorneys and carefully considering our options,” Henderson wrote.
Rachael Dunn, a spokesperson for Missouri Republican Secretary of State Denny Hoskins, wrote in an email that the state hadn’t entered into an agreement with the Justice Department “at this time.”
DOJ ‘contractor’ could get voter data
The draft agreement would give the Justice Department wide authority to share the voter data of states that sign on.
The department would be authorized to share the data with “a contractor” who needs access “to perform duties related” to voter list maintenance verification, according to the draft agreement. The agreement doesn’t name any contractors or specify whether they would be inside or outside of government.
Two states have publicly rejected the draft agreement. Colorado Democratic Secretary of State Jena Griswold announced Dec. 3 she would refuse to sign the memorandum. The Justice Department later sued Colorado.
The Wisconsin Elections Commission also rejected the draft agreement that week. In a Dec. 11 letter to Neff, the Justice Department official, the commissioners wrote that state law prohibits them from releasing certain personally identifiable information, such as date of birth, Social Security numbers and driver’s license numbers.
“I don’t look at the action that we’re taking today to be commentary on the motive of the appropriateness of the Department of Justice’s request,” Commissioner Don Millis, a Republican appointee, said at a virtual commission meeting the same day. “The U.S. DOJ is simply asking the commission to do something that the commission is explicitly forbidden by Wisconsin law to do.”
The Justice Department on Thursday sued Wisconsin for its voter data.
Justin Levitt, who served as senior policy adviser for democracy and voting rights in the Biden White House and is now a law professor at Loyola Marymount University, told Stateline in an email that he expects no states to sign the agreement.
“It’s no surprise that both Colorado and Wisconsin said no — and I don’t think that’s a question of political leadership,” Levitt wrote. “It’s hard for me to imagine any Republican state with faith in its own list maintenance capacity agreeing to outsource that decision to the DOJ.”
Editor’s note: This story has been updated to correct when Colorado refused to sign the memorandum and when the suit was brought against the state. Colorado Newsline’s Lindsey Toomer contributed reporting. Stateline reporter Jonathan Shorman can be reached at jshorman@stateline.org.
This story was originally produced by Stateline, which is part of States Newsroom, a nonprofit news network which includes Wisconsin Examiner, and is supported by grants and a coalition of donors as a 501c(3) public charity.
Federal and local law enforcement officers arrest a man in Washington, D.C., in August. The number of arrests nationwide fell sharply in 2020 and have stayed down since then, according to a new report from the nonpartisan think tank Council on Criminal Justice. (Photo by Andrew Leyden/Getty Images)
Arrests in the United States have fallen to levels not seen in decades, according to a new report that reconstructs national arrest trends in the absence of federal data.
The Council on Criminal Justice, a nonpartisan think tank, on Thursday released the first comprehensive national analysis of arrests since federal authorities stopped publishing detailed arrest statistics in 2020.
Arrests plunged during the first year of the pandemic and have remained low, according to the analysis. The national arrest rate in 2024 was 30% below the 2019 level and 71% lower than the peak in 1994.
Drug arrests have fallen even faster, with adult and juvenile drug-offense arrest rates dropping to about half of what they were in 2019.
In 1980, juveniles made up nearly a fifth of arrests nationwide, but by 2018, their share had fallen to 7%. While adult arrest rates declined 7% between 2020 and 2024, juvenile rates rose 14% over the same period.
Gender patterns have shifted as well. With arrests of men falling more steeply over time, women now account for a larger portion of arrests. Adult women’s share nearly doubled between 1980 and 2020, rising from 14% to about 27%. Girls’ share of juvenile arrests grew from 18% to roughly 31%.
Between 2020 and 2024, arrest rates for Black and Asian juveniles surged 48% and 45%, respectively, compared with an 11% increase among white youth. Rates for American Indian and Alaska Native juveniles fell 4%.
Among adults, arrest rates increased by 12% for Black people and 18% for Asian people, but declined by 10% for white adults and 17% for American Indian and Alaska Native adults.
This story was originally produced by Stateline, which is part of States Newsroom, a nonprofit news network which includes Wisconsin Examiner, and is supported by grants and a coalition of donors as a 501c(3) public charity.
A Delray Beach, Fla., police officer speaks with a driver in 2019. The Trump administration wants access to state driver’s license data through a computer network used by law enforcement to share records across state lines. (Photo by Joe Raedle/Getty Images)
Four Republican states have agreed to help the Trump administration gain access to state driver’s license data through a nationwide law enforcement computer network as part of the administration’s hunt for alleged noncitizen voters.
The Trump administration said as recently as October that federal officials wanted to obtain driver’s license records through the network.
The commitment from officials in Florida, Indiana, Iowa and Ohio comes as part of a settlement agreement filed on Friday in a federal lawsuit. The lawsuit was originally brought by the states last year alleging the Biden administration wasn’t doing enough to help states verify voter eligibility.
The settlement, between the states and the U.S. Department of Homeland Security, requires the federal department to continue its development of a powerful citizenship verification program known as SAVE. Earlier this year, federal officials repurposed SAVE into a program capable of scanning millions of state voter records for instances of noncitizen registered voters.
In return, the states have agreed to support Homeland Security’s efforts to access the National Law Enforcement Telecommunications System, an obscure computer network that typically allows law enforcement agencies to search driver’s license records across state lines. Nlets — as the system is known — lets police officers easily look up the driving records of out-of-state motorists.
The Trump administration and some Republican election officials have promoted the changes to SAVE as a useful tool to identify potential noncitizen voters, and Indiana had already agreed to provide voter records. Critics, including some Democrats, say the Trump administration is building a massive database of U.S. residents that President Donald Trump or a future president could use for spying or targeting political enemies.
Stateline reported last week, before the settlement agreement was filed in court, that Homeland Security publicly confirmed it wants to connect Nlets to SAVE.
A notice published Oct. 31 in the Federal Register said driver’s licenses are the most widely used form of identification, and that by working with states and national agencies, including Nlets, “SAVE will use driver’s license and state identification card numbers to check and confirm identity information.”
A federal official also previously told a virtual meeting of state election officials in May that Homeland Security was seeking “to avoid having to connect to 50 state databases” and wanted a “simpler solution,” such as Nlets, according to government records published by the transparency group American Oversight.
The new settlement lays out the timeline for how the Trump administration could acquire the four states’ records.
Within 90 days of the execution of the agreement, the four states may provide Homeland Security with 1,000 randomly selected driver’s license records from their state for verification as part of a quality improvement process for SAVE.
According to the agreement, the states that provide the records will “make best efforts to support and encourage DHS’s efforts to receive and have full use of state driver’s license records from the National Law Enforcement Telecommunications System” and state driver’s license agencies.
The language in the agreement is open-ended and doesn’t make clear whether the pledge to help Homeland Security obtain access to Nlets is limited to drivers from those four states or is intended to require the states to help the agency acquire the records of drivers nationwide.
An agreement to help
The agreement could pave the way for Republican officials in other states to provide access to license data.
Nlets is a nonprofit organization that facilitates data sharing among law enforcement agencies across state lines. States decide what information to make available through Nlets, and which agencies can access it. That means the four states could try to influence peers to share Nlets data with the Trump administration.
“They’re not just talking about driver’s license numbers, they’re talking about the driver’s records. What possible reason would DHS have in an election or voting context — or any context whatsoever — for obtaining the ‘full use of state driver’s license records,’” said David Becker, executive director of the nonpartisan Center for Election Innovation & Research.
Iowa Secretary of State Paul Pate, a Republican, said in a statement to Stateline that the settlement agreement provides another layer of election integrity and protection as officials seek to ensure only eligible voters are registered. He didn’t directly address questions about Nlets access.
“The SAVE program provides us with critical information, but we must also continue to utilize information from other state and federal partners to maintain clean and accurate lists,” Pate said in the statement.
Two weeks before the Nov. 5, 2024, election, Pate issued guidance to Iowa county auditors to challenge the ballots of 2,176 registered voters who were identified by the secretary of state’s office as potential noncitizens. The voters had reported to the state Department of Transportation or another government entity that they were not U.S. citizens in the past 12 years and went on to register to vote, according to the guidance.
In March, Pate said his office gained access to the SAVE database and found 277 of those people were confirmed to not have U.S. citizenship — just under 12% of the individuals identified as potential noncitizens.
What possible reason would DHS have in an election or voting context — or any context whatsoever — for obtaining the ‘full use of state driver’s license records.’
– David Becker, executive director, Center for Election Innovation & Research
Homeland Security and the U.S. Department of Justice didn’t immediately respond to requests for comment Monday.
Matthew Tragesser, a spokesperson for U.S. Citizenship and Immigration Services — the agency under Homeland Security that oversees SAVE — told Stateline last week that USCIS was committed to “eliminating barriers to securing the nation’s electoral process.”
“By allowing states to efficiently verify voter eligibility, we are reinforcing the principle that America’s elections are reserved exclusively for American citizens,” Tragesser said in a statement.
The SAVE program — Systematic Alien Verification for Entitlements — was originally intended to help state and local officials verify the immigration status of individual noncitizens seeking government benefits. In the past, SAVE could search only one name at a time. Now it can conduct bulk searches; federal officials in May also connected the program to Social Security data.
“It’s a potentially dangerous mix to put driver’s license and Social Security number and date of birth information out there … where we really don’t know yet how and when and where it’s going to be used,” Minnesota Democratic Secretary of State Steve Simon said in an interview on Monday.
Democratic states object
As the Trump administration has encouraged states to use SAVE, the Justice Department has also demanded states provide the department with unredacted copies of their voter rolls. The Trump administration has previously confirmed the Justice Department is sharing voter information with Homeland Security.
The Justice Department has sued six, mostly Democratic, states for refusing to turn over the data. Those lawsuits remain pending.
On Monday, 12 state secretaries of state submitted a 29-page public comment, in response to SAVE’s Federal Register notice, criticizing the overhaul. The secretaries wrote that while Homeland Security claims the changes make the program an effective tool for verifying voters, the modifications are “likely to degrade, not enhance” states’ efforts to ensure free, fair and secure elections.
“What the modified system will do … is allow the federal government to capture sensitive data on hundreds of millions of voters nationwide and distribute that information as it sees fit,” the secretaries wrote.
The secretaries of state of California, Connecticut, Massachusetts, Maine, Michigan, Minnesota, New Jersey, New Mexico, Oregon, Rhode Island, Vermont and Washington signed on to the comment.
The settlement agreement purports to make this year’s changes to SAVE legally binding.
The agreement asks that a federal court retain jurisdiction over the case for 20 years for the purposes of enforcing it — a move that in theory could make it harder for a future Democratic president to reverse the changes to SAVE.
But Becker, of the Center for Election Innovation & Research, said he doesn’t expect the settlement agreement would make it more difficult for a future administration to undo the overhaul.
“Should a different administration come in that disagrees with this approach,” Becker said, “I would expect that they would almost certainly completely change how the system operates and how the states can access it and what data the federal government procures.”
Iowa Capital Dispatch reporter Robin Opsahl contributed to this report. Stateline reporter Jonathan Shorman can be reached at jshorman@stateline.org.
This story was originally produced by Stateline, which is part of States Newsroom, a nonprofit news network which includes Wisconsin Examiner, and is supported by grants and a coalition of donors as a 501c(3) public charity.
A California Highway Patrol officer talks to a driver during a traffic stop in October. The U.S. Department of Homeland Security wants access to state driver’s license data as it builds a powerful citizenship verification program. (Photo by Justin Sullivan/Getty Images)
The Trump administration wants access to state driver’s license data on millions of U.S. residents as it builds a powerful citizenship verification program amid its clampdown on voter fraud and illegal immigration.
The U.S. Department of Homeland Security seeks access to an obscure computer network used by law enforcement agencies, according to a federal notice, potentially allowing officials to bypass negotiating with states for the records.
The information would then be plugged into a Homeland Security program known as SAVE that Trump officials have deployed to search for rare instances of alleged noncitizen voters and to verify citizenship. The plan comes as the Trump administration demands states share copies of their voter files that include sensitive personal data that also is being plugged into SAVE; it is suing some states that refuse.
Trump officials tout the SAVE program as a boost for election integrity. But critics of the program warn the federal government is constructing a massive, centralized information source on Americans. They fear President Donald Trump or a future president could use the tool to surveil residents or target political enemies.
“What this SAVE database expansion will do is serve as a central pillar to build dossiers on all of us,” said Cody Venzke, a senior policy counsel at the American Civil Liberties Union.
At the same time, Homeland Security Investigations and U.S. Immigration and Customs Enforcement, or ICE, conducted nearly 900,000 searches for state driver’s license and other motor vehicle data over the past year using the same data-sharing network that Homeland Security wants to link to SAVE, according to information provided to Congress. The network is called Nlets — formerly the National Law Enforcement Telecommunications System, now known as the International Justice and Public Safety Network.
Dozens of congressional Democrats in mid-November warned Democratic governors that Nlets makes driver’s license data available to ICE, including from states that restrict cooperation with the agency. While ICE, a Homeland Security agency, has long had access to Nlets, some Democrats are voicing renewed alarm amid Trump’s sweeping deportation campaign.
At least five states — Illinois, Massachusetts, Minnesota, New York and Washington — have blocked Nlets’ ability to share their driver’s license records with ICE, according to the Nov. 12 letter from 40 Democratic lawmakers. Oregon also is taking steps to block access.
In Colorado, state Sen. Julie Gonzales said she is willing to advance bills to block the Nlets data sharing. Gonzales, a Democrat who chairs the Colorado Senate Judiciary Committee, has previously sponsored legislation to limit what personal information is shared with the federal government for immigration enforcement.
“It is like playing Whac-A-Mole, but the Constitution applies to ICE, too,” Gonzales said.
The recent developments underscore the ongoing struggle between Democratic states and the Trump administration over how much access Homeland Security should have to their residents’ personal data. For their part, some Republican state officials have voiced support for the administration’s moves and want to aid the search for noncitizen voters and individuals in the country illegally.
Data and privacy experts told Stateline the current moment could lead to more centralization of personal data by the federal government and an eroding expectation of privacy when it comes to driver’s license information. The federal government is for the first time essentially building a U.S. citizenship database, they said.
Homeland Security is proposing to take Nlets outside its intended use, said John Davisson, senior counsel and director of litigation at the Electronic Privacy Information Center, a Washington, D.C.-based research and advocacy group that argues privacy is a fundamental right.
Nlets is a nonprofit organization that facilitates data sharing among law enforcement agencies across state lines. At a basic level, Nlets is the system that allows police officers to quickly look up the driver’s license information of out-of-state motorists they pull over.
States decide what information to make available through Nlets, and which agencies can access it. Each state has an Nlets member, typically that state’s highway patrol or equivalent agency. Several federal law enforcement agencies also are members.
“It appears that DHS is eyeing it for something quite different, for mass extraction of driver’s license information that would be far beyond the sort of targeted enforcement purposes of a system like Nlets,” Davisson said.
Driver data idea floated in May
Homeland Security’s SAVE program — Systematic Alien Verification for Entitlements — was originally intended to help state and local officials verify the immigration status of individual noncitizens seeking government benefits. But it can now scan state voter rolls for alleged noncitizen voters.
In the past, SAVE could search only one name at a time. Now it can conduct bulk searches, allowing officials to potentially scan through information on millions of registered voters. Federal officials in May connected the program to Social Security data; linking driver’s license data through Nlets would provide an additional mountain of data on U.S. residents.
The League of Women Voters, a nonpartisan group that advocates for voting rights, filed a federal lawsuit in September against Homeland Security over the transformation of SAVE. In its complaint, the organization accused the department of ignoring federal law to create comprehensive databases of American citizens’ data.
U.S. District Court Judge Sparkle L. Sooknanan, a Biden appointee, last week declined to temporarily block SAVE’s overhaul while the lawsuit proceeds. But Sooknanan wrote in an opinion that based on the current record, “the Court is troubled by the recent changes to SAVE and doubts the lawfulness of the Government’s actions.”
Homeland Security publicly confirmed it wants to connect Nlets to SAVE in an Oct. 31 Federal Register notice. The notice said driver’s licenses are the most widely used form of identification, and by working with states and national agencies, including Nlets, “SAVE will use driver’s license and state identification card numbers to check and confirm identity information.”
The agency also privately floated its interest in Nlets months earlier.
According to minutes of a May virtual meeting of the National Association of Secretaries of State Elections Committee, U.S. Citizenship and Immigration Services (USCIS) official Brian Broderick told the group that his agency — the Homeland Security agency that administers SAVE — was seeking “to avoid having to connect to 50 state databases” and wanted a “simpler solution,” such as Nlets.
The minutes were contained in records from the Texas Secretary of State’s Office obtained by American Oversight, a nonpartisan transparency group that frequently files records requests. Mother Jones magazine first reported on the records.
Nlets and the Texas Secretary of State’s Office didn’t respond to requests for comment.
On Friday, National Association of Secretaries of State spokesperson Brittany Hamilton wrote in an email to Stateline that at that time, “we have not received specific updates from USCIS on this aspect of driver’s license data potential usage.”
In a statement, USCIS spokesperson Matthew Tragesser encouraged all federal, state and local agencies to use SAVE.
“USCIS remains dedicated to eliminating barriers to securing the nation’s electoral process. By allowing states to efficiently verify voter eligibility, we are reinforcing the principle that America’s elections are reserved exclusively for American citizens,” Tragesser wrote.
State restrictions flawed, lawmakers say
Some Democrats are separately pushing to limit ICE’s access to driver’s license data through Nlets. The Nov. 12 congressional letter warned that while some states have restrictions on data sharing with immigration authorities, the limits are often ineffective because of major flaws.
State limits sometimes apply only to state motor vehicle agencies, which don’t connect to Nlets — and often don’t apply to state police agencies that do connect, the letter said. And even though state restrictions target data-sharing for immigration enforcement, Nlets doesn’t indicate the purpose of a request.
“Because of the technical complexity of Nlets’ system, few state government officials understand how their state is sharing their residents’ data with federal and out-of-state agencies,” wrote U.S. Sen. Ron Wyden of Oregon, U.S. Rep. Adriano Espaillat of New York and 38 other Democrats.
Homeland Security didn’t address Stateline’s questions about ICE’s access to state driver’s license data through Nlets.
I think that for many years, folks around the country that are concerned about privacy, that are concerned about immigrants, have been trying to sound the alarm about this issue.
– Matthew Lopas of the National Immigration Law Center
Advocates for immigrants have long raised concerns about ICE access to state driver’s license data through Nlets. Nineteen states allow residents to obtain driver’s licenses regardless of immigration status, according to the National Immigration Law Center, an immigrant advocacy group. Those driver’s license records represent a wealth of information on noncitizens.
While ICE can’t use Nlets to obtain records of all noncitizens issued licenses, the agency can use the search tool to obtain a variety of information on individuals, such as date of birth, sex, address and Social Security number, according to the law center. Sometimes a photo is also available — a particular concern for immigrants and their advocates amid reports that ICE has deployed facial recognition tools in the field.
“I think that for many years, folks around the country that are concerned about privacy, that are concerned about immigrants, have been trying to sound the alarm about this issue,” said Matthew Lopas, director of state advocacy and technical assistance at the National Immigration Law Center.
Stateline contacted all 50 state governors to ask about Nlets. Forty-one offices didn’t respond and most others provided high-level statements or referred questions to other agencies.
But Maryland indicated it was taking “proactive measures” to ensure that federal agencies’ access to its data through Nlets complies with state and federal law. A 2021 state law limits the sharing of driver’s license data with federal immigration authorities.
Maryland “is working with Nlets to ensure that Marylanders’ data is not misused for civil immigration enforcement absent a valid judicial warrant, and we intend to share more information on that effort as we are able,” Rhyan Lake, a spokesperson for Maryland Democratic Gov. Wes Moore, said in a statement to Stateline.
The South Dakota Department of Public Safety, which is overseen by Republican Gov. Larry Rhoden, cautioned against limiting data-sharing among law enforcement. Records obtained through Nlets include data on wanted individuals and other information that can help identify potential threats to officers and agents, the department said in a statement provided by Director of Communications Brad Reiners.
“We reject the concerns outlined in the [Democratic lawmakers’] letter and remain deeply concerned about the potentially dangerous consequences of limiting access to this information,” the statement says.
In Oregon, state officials plan to cut off ICE’s Nlets access to its driver’s license data, but no date has been set, Oregon State Police Capt. Kyle Kennedy, an agency spokesperson, wrote in an email.
“We are working with other states to assist in considering a path forward,” Kennedy wrote.
This story was originally produced by Stateline, which is part of States Newsroom, a nonprofit news network which includes Wisconsin Examiner, and is supported by grants and a coalition of donors as a 501c(3) public charity.
After a year of being STN’s Transportation Director of the Year, Craig Beaver of Beaverton School District in Oregon joins us to discuss the ins and outs of running a large mixed-fleet school bus operation, pushing the limits with technology and data, navigating current federal changes, and looking to the future of the industry.
This is Hongyu Liu, Wisconsin Watch’s new data investigative reporter.
If you’ve ever been confused and even intimidated by statistics and other numbers, I feel you.
I was in the same boat three years ago while interning at a newspaper in Quincy, Massachusetts.
Wisconsin Watch data investigative reporter Hongyu Liu (Joe Timmerman / Wisconsin Watch)
When gas prices soared, my colleagues and I made weekly calls to every gas station in town, asking for price updates. It was an effort to help readers who were not familiar with gas price apps to learn where to fill up their cars more affordably. But we were soon drowning in data. Each week, we posted a lengthy list of individual prices, which were already one day old when reaching the readers’ doorsteps. We didn’t quite know how to look at the numbers in a more thoughtful, useful way.
Had I the analysis skills I’ve since developed, I would have approached the assignment differently. I would have looked for trends that may have inspired stories about how the higher gas prices might tighten the budgets of residents.
My eagerness for understanding the world of numbers prompted me to pursue a master’s degree in data journalism at Columbia University. There, I found my niche is where data analysis, web design and journalistic storytelling intersect. I went on to spend almost two years in Charleston, South Carolina, as a data reporting fellow at The Post and Courier, becoming the newsroom “data nerd.” I used data skills to sift through drug prescription records to find evidence of identity theft and understand how loosened regulations led to a surge in sea turtle deaths from dredging near ports.
Now I’m eager to do similar reporting for Wisconsin, using data to provide rich context to our journalism that aims to make communities strong, informed and connected. That includes finding investigative leads rooted in data and producing visualizations that explain the issues we cover, such as through our DataWatch series.
We live in a world with ever-increasing reams of raw data that, if understood and analyzed, can help us better understand our communities. I’m stepping in at Wisconsin Watch to take the lead on how we use data in our journalism — and to understand how actors across the state are representing or even misrepresenting data trends.
I want to hear from you. If you have ideas for data we should analyze and visualize, or if you have questions about data in a government report, email me at hliu@wisconsinwatch.org and share your thoughts.
Wisconsin Watch is a nonprofit, nonpartisan newsroom. Subscribe to our newsletters for original stories and our Friday news roundup.
Now that you’ve made it through the beginning of the new school year, this is the optimal time to tune your transportation operations. Learn how to leverage your live ridership, routing, and call volume data to reduce missed stops, decrease parent inquiries, and enhance on-time performance.
Join Pathwise and School Transportation News on Thursday, October 9, at 10:00 a.m. PT / 1:00 p.m. ET for a live 60-minute session to discover how to convert your early-semester data into concrete mid-year gains without replacing your current routing platform.
In this session, you’ll get practical tips on how to:
Reassess your routes using current ridership signals (scan data, driver logs, parent app activity, no-show patterns)
Improve routing reliability with targeted fixes for tier balancing, stop consolidation, bell-time alignment, and more, instead of major re-routes
Track ridership more accurately using count audits, exception workflows, and reconciliations, and turn these insights into schedule improvements
Operationalize KPIs that matter—on-time percentage, call-center volume, and parent notification latency—so you know where adjustments may need to be made.
Bottom line: This isn’t starting over; it’s making smarter use of tools, data, and processes you already have—to ensure smoother operations throughout the rest of the year.
Who should attend: Transportation directors, routing/dispatch leads, and operations managers.
Brought to you by Pathwise
REGISTER BELOW:
Presenters:
Michael Roche
VP of Customer Engagement and Business Development
EZRouting
With over 13 years of experience as a Director of Transportation for a school district, Roche possesses extensive expertise in overseeing logistical operations and ensuring the safety and efficiency of transportation systems. Transitioning into consulting, he has utilized his knowledge to aid school districts in optimizing transportation operations and implementing software solutions. Currently, Roche is committed to collaborating with school districts across the country, assisting them in maximizing the benefits of the software and providing comprehensive consulting services tailored to their transportation requirements.
Carl Allen
Chief Executive Officer
4MATIV
Carl Allen is an experienced leader in education, transportation, and public policy, currently serving as CEO and founder of 4MATIV Technologies, which he launched in 2018. He previously served as Director of Transportation for Boston Public Schools, Regional Vice President for Transdev in Colorado, and COO/CFO of a charter school network in Minneapolis. Drawing on his training in urban planning and public policy from Harvard’s Kennedy School, and his early experience as a Peace Corps Volunteer teaching high school math in Ghana, Allen supports school districts in tackling complex transportation challenges. He holds degrees in industrial and manufacturing design engineering from Northwestern University and lives in St. Paul, Minnesota with his wife and three children.
Hundreds of thousands of students are on new routes to and from school this month.
While some school districts may still be tracking these numbers manually, many
transportation departments are implementing new technology to take the guesswork out of student ridership.
Luisa Brown is wearing two hats at Zillah School District in Washington, that of an accounts payable supervisor and transportation manager. When she started in the latter role in March 2020, and without a long background in student transportation, she leaned heavily on technology for all the assistance she could get.
Brown said that despite working at a smaller school district that transports approximately 662 students daily, she realized that tracking routing via spreadsheets was not an ideal solution. That’s when she first started using the Tyler Technologies routing software, implemented in December 2020. The student ridership verification
technology via RFID student cards was added in 2023.
A phased approach to implementing new technology was necessary from a budgetary standpoint, she noted, which also was essential for ensuring the technology is utilized correctly and benefitting the student transportation staff.
Tim Ammon, a consultant in the student transportation industry since 2001, said the “Holy Grail” of this kind of technology is the amount of intervention required.
Ammon explained that in his experience as a consultant and working in the business management of school bus technology (he recently served as VP and GM of passenger services for Zonar Systems and remains a strategic advisor), he sees two main uses of student ridership verification.
“The first is, in the event that something goes wrong, we can track back to where the kid got on and off the bus and at least have a starting point. So, emergency district management applications.”
In Brown’s case, integration was smooth, since she said she was already using Tyler’s routing software and Tyler Drive to connect with the RFID cards. But in Colorado, Denver Public Schools (DPS) ran into challenges as transportation prepared to roll out student ridership technology last month for the first time.
“Samsara has been a very willing and helpful partner in making sure all the components of our project roll-out smoothly and are operational internally,” said Tyler Maybee, director of operations for Denver’s transportation services, who said the district is creating an in-house student ridership technology solution with the GPS provider alongside a smaller technology company.
“We have another vendor that is more of a barrier than opportunistic and has prevented our innovation from raising the bar within their own technology. It has forced us to find many workarounds and begin to search for a better partner that has a similar vision to fully integrate transportation technology.”
With about 5,000 to 7,000 students being transported daily across Denver, Maybee said time will tell the success of the new project.
“But all signs point to a more knowledgeable and connected DPS community and a reduction in the number of calls our dispatchers receive regarding missing students and requests for bus information,” he said.
Keeping Data Secure
On the topic of data security for this type of technology, Ammon noted it’s crucial to have “procedural aspects in place to make sure that you know that information is
protected.” Easier said than done as it’s a process that can have an “enormous number of tentacles into it,” he added.
An Education Week article found that education was the fourth-most targeted sector during the first half of 2025, based on data collected by Comparitech.
“Schools are tempting targets for hackers because they have tons of sensitive data and have become more reliant than ever on digital tools,” the article stated. Amy McLaughlin, the project director for the Consortium for School Networking’s (CoSn) cybersecurity initiatives, was quoted saying that districts are aware of the security concerns but face challenges of funding and staff to ensure that data and cybersecurity issues are adequately addressed.
Brown said she keeps physical security on a tight lockdown as each tablet has a unique PIN that only she and the individual driver has access to.
Bill Westerman, Tyler’s director of integration solutions, confirmed that all Tyler Drive tablets are encrypted and that districts can choose how registration information is shown when student data is being inputted.
Maybee said the Denver IT team has a series of regulations in place to prevent student data from falling into the wrong hands and that vendors are required to sign a data privacy agreement “to make sure their systems meet the same level of security our network has to maintain adequate protections,” he continued. “We limited the amount of personal identifiable information on the ID virtual and physical ID cards to make sure even if a card was misplaced and then subsequently found that a student’s information is not at risk. This also includes encrypting the QR code so that a scan must be tied to our system to make any sense out of the resulting scan data.”
Edulog’s Lam-Nyugen Bull, who serves as the company’s chief experience officer, said the software company maintains SOC 2, Type 2 compliance and that “all data is encrypted at rest and in transit and we regularly undergo third-party penetration testing and evaluation of our overall security posture.”
As a certified risk manager, Ammon encouraged student transportation professionals to find resources or individuals that can assist with being able to “talk to your vendor intelligently about their data security procedures.”
Especially when integrating different vendors’ technology options into one transportation operation, he said that collaboration is crucial with increased risk of
malicious cybersecurity attacks.
“From a vendor perspective, it’s very likely that each district will have its own flavor of how it wants to deal with this, and so like as a vendor, I should know that,
right? Because I should be responding to what your requirements are as a customer, right? To assume that all 16,000 school districts in the country want exactly the same response in the event of it is, I think, a fallacy,” Ammon said. “There should be some collaboration between the district and the vendor in terms of, here’s our expectations around this, here’s the universe of what’s possible. How do we want to narrow that universe so that it fits whatever we’re doing?”
Evolving Technology
RFID cards, QR codes, barcodes and manual checklists are all ways that student ridership can be documented. Most industry experts agree that RFID cards can help keep tabs on the students on the bus without exposing their information, but what are the future possibilities when it comes to this technology?
Ammon noted that video camera facial recognition or biometric scans are trickier territory to navigate as those types of technology naturally raise a high level of privacy concerns with parents.
“There is no technology impediment today that would stop us from doing [options like biometric scanning],” said Zach Moren, Transfinder’s manager of sales enablement and engineering. “But schools need to consider a few things when looking at ridership solutions. What is the most cost effective? What is the most reliable to capture as close to 100 percent of riders as possible? And what technology can be easily adopted and utilized by bus drivers, students and the community? Based on those requirements I’m skeptical we will see a major change in technology anytime in the near future because RFID solves each of those challenges so effectively.”
Moren noted that Transfinder is developing a digital wallet card that students could access on their smartphones, “like they would a credit card or concert ticket,” which Moren said could address the issue of RFID cards being lost or damaged.
“As schools continue to prioritize student well-being, the evolution of ridership verification technology is set to move beyond isolated solutions and adopt a more holistic approach, intertwining safety and health measures with the core mission of ensuring every child’s secure passage to and from school,” said Edulog’s Nyugen-Bull
when discussing the future of this technology.
Brown noted that one Tyler software feature she found to be immensely helpful is the ability to run health reports to make sure drivers were aware of health information for the students on their routes, such as food allergies or other relevant factors such as anxiety. She said this information was historically kept in a folder or backpack on the bus, which was not the best way for drivers to quickly access the information and be aware of important student information or emergency contact details.
She also noted that Tyler is doing “an amazing job of making updates throughout the year, so that it’s not just a dead program and [it’s] improving every year,” she continued. “And I think they do an amazing job in getting the in-user’s input because they are creating something that they feel is going to work for everybody.
Because there [are] different circumstances in small districts versus large districts.”
Integration and collaboration continue to be important factor for companies and districts as they work together to keep student data secure and improve on the implementation of this technology to benefit not only the students but student transportation operational workings.
Editor’s Note: As reprinted from the September 2025 issue of School Transportation News.
Increasingly, the conversation about cybersecurity and data protection includes student transportation. STN addressed the subject of security in the September magazine issue, featuring articles that focused on video camera storage and security as well as data security and routing.
STN spoke with Jake McOmie, the CTO of Confluence Security, a systems integrator company that brings together products from various manufacturers — of cameras, recording devices, servers, networking equipment, and sensors — to create tailored security systems. These systems are designed to address both physical and cybersecurity needs with an emphasis on automation, identity management and analytics. The company, which works with government, school and commercial or enterprise customers, also provides software that unifies all components, enabling features like real-time alerts, video analytics and automated response to security events.
STN: Why is security and cybersecurity important for school districts and transportation departments right now?
McOmie: Security and cybersecurity aren’t new concerns, but in today’s connected world, they are more critical than ever. School districts are rapidly adopting technologies like IP cameras, GPS systems, Wi-Fi routers and student tracking software. These tools improve safety and efficiency, but each device added to the network also introduces potential vulnerabilities.
We call this security of security, a phrase borrowed from our trusted manufacturer partner of open-architecture security software platform, Genetec. The approach ensures a cybersecurity-first posture and it’s critical practice to understand your product choices are being systemically protected by design, not as an afterthought.
How concerned are you about the data security of your student transportation operations?
In the age of the Internet of Things (IoT), everything is interconnected. One unsecure device — whether a camera, HVAC sensor, or access control point — can act as the weak link that compromises the entire system. No matter how robust a network may be, its strength depends on every component being secure. That’s why it’s not enough to harden just the network. Districts must vet the products themselves, hold manufacturers accountable for cybersecurity practices and ensure every piece of technology is built with a “security-first” mindset.
Trust is earned, not assumed. Cybersecurity must be woven into procurement, deployment and management. When one compromised camera or device can become an open door, due diligence isn’t optional. It’s essential.
STN: How can transportation departments ensure their data is protected? What steps should they be taking?
Jake McOmie, CTO of Confluence Security
McOmie: Transportation departments manage highly sensitive data, including student info, vehicle locations, incident videos and operational logs. To protect this data, a comprehensive approach during the initial planning will ensure this sensitive data is not jeopardized from unauthorized access. We can talk about the various aspects end users should keep forefront during the planning phase
Vet manufacturers and integrators. Work only with vendors that prioritize cybersecurity and provide transparent security documentation. Vendors who operate under zero-trust security policies and demand nothing less of their technology partners, should be asked early in the process. It’s a pass or fail question and should be enforced without hesitation.
Network segmentation. Isolate transportation and security systems from general-use school networks. Implementing advanced enterprise segmentation through Federations allows for controlled third-party access while maintaining autonomous and isolated authorization. Preferably utilize SaaS-hosted federation services so partner agencies, such as between schools and 911 centers, can connect their networks for data sharing without actually connecting to anything except the mediary cloud-hosted federation server. This method adds the benefit of permission-based access at the most minute level of data, like allowing access to a video feed only if three independent trigger points have verified.
Multi-factor authentication (MFA). Implement MFA at all levels — application logins, device portals and cloud platforms — to prevent account takeovers, especially when passwords are compromised.
Zero-trust approach. Assume no device or user is secure by default. Require verification and limit access by role. To maximize the effects of this policy, utilize automations and/or integrations to minimize the number of touchpoints when permission changes occur.
Encryption & updates. Use end-to-end encryption for data in motion and ensure firmware/software is routinely patched. If available, consider using SaaS products to perform all or some tasks, which can help protect systems from becoming outdated, even if only for a short duration.
Automation & alerting. Leverage tools that can automatically identify patterns or anomalies and escalate issues to the right personnel. Open-architecture systems allow for a larger variety of inputs, and with proper configuration, the sensors can be associated with other sensors or events to help qualify any given scenario before notifying personnel, and ensure the correct personnel are the ones being notified.
Protecting data is not just about prevention. It’s about building resilience and ensuring your team can respond quickly and effectively when an event occurs.
STN: How do you advise school districts to work with their technology department?
McOmie: One of the most common challenges we see is operational silos. Safety and security departments know the problems they need to solve, but IT departments hold the keys to implementation. Successful projects require early and continuous collaboration between these teams.
At Confluence Security, we provide end-to-end IP-based solutions, which means we’re deeply engaged with IT teams during planning, design and deployment. While safety leaders define the why, IT ensures the how is executed securely and effectively. The IT team is critical in achieving a successfully hardened system and should include these three key points:
Designing the network architecture to limit exposure.
Setting access controls and firewall rules.
Validating compliance with cybersecurity policies.
In today’s world, a zero-trust model is no longer optional. Every actor, internal or external, must be authenticated and authorized. School districts can support this by standardizing processes like MFA and ensuring IT reviews any new connected hardware or software before it’s deployed.
STN: Where do you see AI in security?
McOmie: AI is transforming security in two important ways — behind the scenes and in front of the user.
Behind the scenes, AI helps devices self-optimize — learning traffic patterns, refining video compression, or detecting performance anomalies before they become problems. This isn’t flashy, but it’s foundational to deliver faster, smarter, more reliable systems. The increased accuracy and performance is generally appreciated by end users but in today’s world of tech, the continual improvements are more or less expected.
Video Analytics engines, where video streams are computer-analyzed for specific behaviors, have used AI to improve their intelligence for more than a decade in some cases. In this method, software developers gain tremendous assistance with perfecting their analytical algorithms. In recent years, advancemnts have been made so far as to providing users with the ability to generate their own behavior definitions and AI creates the behavior analysis, delivering a DIY approach to video analytics.
From the user perspective, AI enhances how we interact with security systems. Instead of digging through hours of video, users can issue simple commands: “Show me anything unusual at Bus Lot A last night,” or “Search for students wearing red backpacks on buses 12 thru 15 last week.”
AI enables faster investigations and richer situational awareness. Rather than responding to noise (e.g., constant motion alerts), users receive qualified insights based on anomalies — events that stand out from the norm, like a student jumping out of an open bus window, or a person loitering in an atypical location.
But AI doesn’t stop at behavioral detection. It fundamentally supports action through automation. Systems can support users through if/then/else conditional logic decision making to promote accuracy in the users actions and response. Ultimately, the preferred outcome can be guided by digitized SOPs, allowing for a newbie operator to respond the same way a well-seasoned operator would.
These layers of logic ensure that when serious threats arise, escalation to law enforcement or 911 is intentional, not a false alarm, and delivers real actionable video, data and evidence.
School bus operations rely on technology to address the complex requirements of transporting students. Properly identifying your organization’s operational, functional, technical and financial needs will enhance its capabilities as well as your satisfaction with your choices—but how to start?
Download our complimentary white paper for fresh perspective into choosing a partner who provides good value, not just a good price.
Take a wide, objective look at your organization to understand what’s needed.
Identify the new technologies’ impact to end users and other departments.
Work with potential partners to define KPIs and calculate projected ROI.
Evaluate providers in detail to determine their suitability as a long-term partner.
Ensure regulatory compliance, and look for partnerships and integrations.
Fill out the form below and then check your email for the white paper download link.
On May 6, MIT AgeLab’s Advanced Vehicle Technology (AVT) Consortium, part of the MIT Center for Transportation and Logistics, celebrated 10 years of its global academic-industry collaboration. AVT was founded with the aim of developing new data that contribute to automotive manufacturers, suppliers, and insurers’ real-world understanding of how drivers use and respond to increasingly sophisticated vehicle technologies, such as assistive and automated driving, while accelerating the applied insight needed to advance design and development. The celebration event brought together stakeholders from across the industry for a set of keynote addresses and panel discussions on critical topics significant to the industry and its future, including artificial intelligence, automotive technology, collision repair, consumer behavior, sustainability, vehicle safety policy, and global competitiveness.
Bryan Reimer, founder and co-director of the AVT Consortium, opened the event by remarking that over the decade AVT has collected hundreds of terabytes of data, presented and discussed research with its over 25 member organizations, supported members’ strategic and policy initiatives, published select outcomes, and built AVT into a global influencer with tremendous impact in the automotive industry. He noted that current opportunities and challenges for the industry include distracted driving, a lack of consumer trust and concerns around transparency in assistive and automated driving features, and high consumer expectations for vehicle technology, safety, and affordability. How will industry respond? Major players in attendance weighed in.
In a powerful exchange on vehicle safety regulation, John Bozzella, president and CEO of the Alliance for Automotive Innovation, and Mark Rosekind, former chief safety innovation officer of Zoox, former administrator of the National Highway Traffic Safety Administration, and former member of the National Transportation Safety Board, challenged industry and government to adopt a more strategic, data-driven, and collaborative approach to safety. They asserted that regulation must evolve alongside innovation, not lag behind it by decades. Appealing to the automakers in attendance, Bozzella cited the success of voluntary commitments on automatic emergency braking as a model for future progress. “That’s a way to do something important and impactful ahead of regulation.” They advocated for shared data platforms, anonymous reporting, and a common regulatory vision that sets safety baselines while allowing room for experimentation. The 40,000 annual road fatalities demand urgency — what’s needed is a move away from tactical fixes and toward a systemic safety strategy. “Safety delayed is safety denied,” Rosekind stated. “Tell me how you’re going to improve safety. Let’s be explicit.”
Drawing inspiration from aviation’s exemplary safety record, Kathy Abbott, chief scientific and technical advisor for the Federal Aviation Administration, pointed to a culture of rigorous regulation, continuous improvement, and cross-sectoral data sharing. Aviation’s model, built on highly trained personnel and strict predictability standards, contrasts sharply with the fragmented approach in the automotive industry. The keynote emphasized that a foundation of safety culture — one that recognizes that technological ability alone isn’t justification for deployment — must guide the auto industry forward. Just as aviation doesn’t equate absence of failure with success, vehicle safety must be measured holistically and proactively.
With assistive and automated driving top of mind in the industry, Pete Bigelow of Automotive News offered a pragmatic diagnosis. With companies like Ford and Volkswagen stepping back from full autonomy projects like Argo AI, the industry is now focused on Level 2 and 3 technologies, which refer to assisted and automated driving, respectively. Tesla, GM, and Mercedes are experimenting with subscription models for driver assistance systems, yet consumer confusion remains high. JD Power reports that many drivers do not grasp the differences between L2 and L2+, or whether these technologies offer safety or convenience features. Safety benefits have yet to manifest in reduced traffic deaths, which have risen by 20 percent since 2020. The recurring challenge: L3 systems demand that human drivers take over during technical difficulties, despite driver disengagement being their primary benefit, potentially worsening outcomes. Bigelow cited a quote from Bryan Reimer as one of the best he’s received in his career: “Level 3 systems are an engineer’s dream and a plaintiff attorney’s next yacht,” highlighting the legal and design complexity of systems that demand handoffs between machine and human.
In terms of the impact of AI on the automotive industry, Mauricio Muñoz, senior research engineer at AI Sweden, underscored that despite AI’s transformative potential, the automotive industry cannot rely on general AI megatrends to solve domain-specific challenges. While landmark achievements like AlphaFold demonstrate AI’s prowess, automotive applications require domain expertise, data sovereignty, and targeted collaboration. Energy constraints, data firewalls, and the high costs of AI infrastructure all pose limitations, making it critical that companies fund purpose-driven research that can reduce costs and improve implementation fidelity. Muñoz warned that while excitement abounds — with some predicting artificial superintelligence by 2028 — real progress demands organizational alignment and a deep understanding of the automotive context, not just computational power.
Turning the focus to consumers, a collision repair panel drawing Richard Billyeald from Thatcham Research, Hami Ebrahimi from Caliber Collision, and Mike Nelson from Nelson Law explored the unintended consequences of vehicle technology advances: spiraling repair costs, labor shortages, and a lack of repairability standards. Panelists warned that even minor repairs for advanced vehicles now require costly and complex sensor recalibrations — compounded by inconsistent manufacturer guidance and no clear consumer alerts when systems are out of calibration. The panel called for greater standardization, consumer education, and repair-friendly design. As insurance premiums climb and more people forgo insurance claims, the lack of coordination between automakers, regulators, and service providers threatens consumer safety and undermines trust. The group warned that until Level 2 systems function reliably and affordably, moving toward Level 3 autonomy is premature and risky.
While the repair panel emphasized today’s urgent challenges, other speakers looked to the future. Honda’s Ryan Harty, for example, highlighted the company’s aggressive push toward sustainability and safety. Honda aims for zero environmental impact and zero traffic fatalities, with plans to be 100 percent electric by 2040 and to lead in energy storage and clean power integration. The company has developed tools to coach young drivers and is investing in charging infrastructure, grid-aware battery usage, and green hydrogen storage. “What consumers buy in the market dictates what the manufacturers make,” Harty noted, underscoring the importance of aligning product strategy with user demand and environmental responsibility. He stressed that manufacturers can only decarbonize as fast as the industry allows, and emphasized the need to shift from cost-based to life-cycle-based product strategies.
Finally, a panel involving Laura Chace of ITS America, Jon Demerly of Qualcomm, Brad Stertz of Audi/VW Group, and Anant Thaker of Aptiv covered the near-, mid-, and long-term future of vehicle technology. Panelists emphasized that consumer expectations, infrastructure investment, and regulatory modernization must evolve together. Despite record bicycle fatality rates and persistent distracted driving, features like school bus detection and stop sign alerts remain underutilized due to skepticism and cost. Panelists stressed that we must design systems for proactive safety rather than reactive response. The slow integration of digital infrastructure — sensors, edge computing, data analytics — stems not only from technical hurdles, but procurement and policy challenges as well.
Reimer concluded the event by urging industry leaders to re-center the consumer in all conversations — from affordability to maintenance and repair. With the rising costs of ownership, growing gaps in trust in technology, and misalignment between innovation and consumer value, the future of mobility depends on rebuilding trust and reshaping industry economics. He called for global collaboration, greater standardization, and transparent innovation that consumers can understand and afford. He highlighted that global competitiveness and public safety both hang in the balance. As Reimer noted, “success will come through partnerships” — between industry, academia, and government — that work toward shared investment, cultural change, and a collective willingness to prioritize the public good.
Car design is an iterative and proprietary process. Carmakers can spend several years on the design phase for a car, tweaking 3D forms in simulations before building out the most promising designs for physical testing. The details and specs of these tests, including the aerodynamics of a given car design, are typically not made public. Significant advances in performance, such as in fuel efficiency or electric vehicle range, can therefore be slow and siloed from company to company.
MIT engineers say that the search for better car designs can speed up exponentially with the use of generative artificial intelligence tools that can plow through huge amounts of data in seconds and find connections to generate a novel design. While such AI tools exist, the data they would need to learn from have not been available, at least in any sort of accessible, centralized form.
But now, the engineers have made just such a dataset available to the public for the first time. Dubbed DrivAerNet++, the dataset encompasses more than 8,000 car designs, which the engineers generated based on the most common types of cars in the world today. Each design is represented in 3D form and includes information on the car’s aerodynamics — the way air would flow around a given design, based on simulations of fluid dynamics that the group carried out for each design.
Each of the dataset’s 8,000 designs is available in several representations, such as mesh, point cloud, or a simple list of the design’s parameters and dimensions. As such, the dataset can be used by different AI models that are tuned to process data in a particular modality.
DrivAerNet++ is the largest open-source dataset for car aerodynamics that has been developed to date. The engineers envision it being used as an extensive library of realistic car designs, with detailed aerodynamics data that can be used to quickly train any AI model. These models can then just as quickly generate novel designs that could potentially lead to more fuel-efficient cars and electric vehicles with longer range, in a fraction of the time that it takes the automotive industry today.
“This dataset lays the foundation for the next generation of AI applications in engineering, promoting efficient design processes, cutting R&D costs, and driving advancements toward a more sustainable automotive future,” says Mohamed Elrefaie, a mechanical engineering graduate student at MIT.
Elrefaie and his colleagues will present a paper detailing the new dataset, and AI methods that could be applied to it, at the NeurIPS conference in December. His co-authors are Faez Ahmed, assistant professor of mechanical engineering at MIT, along with Angela Dai, associate professor of computer science at the Technical University of Munich, and Florin Marar of BETA CAE Systems.
Filling the data gap
Ahmed leads the Design Computation and Digital Engineering Lab (DeCoDE) at MIT, where his group explores ways in which AI and machine-learning tools can be used to enhance the design of complex engineering systems and products, including car technology.
“Often when designing a car, the forward process is so expensive that manufacturers can only tweak a car a little bit from one version to the next,” Ahmed says. “But if you have larger datasets where you know the performance of each design, now you can train machine-learning models to iterate fast so you are more likely to get a better design.”
And speed, particularly for advancing car technology, is particularly pressing now.
“This is the best time for accelerating car innovations, as automobiles are one of the largest polluters in the world, and the faster we can shave off that contribution, the more we can help the climate,” Elrefaie says.
In looking at the process of new car design, the researchers found that, while there are AI models that could crank through many car designs to generate optimal designs, the car data that is actually available is limited. Some researchers had previously assembled small datasets of simulated car designs, while car manufacturers rarely release the specs of the actual designs they explore, test, and ultimately manufacture.
The team sought to fill the data gap, particularly with respect to a car’s aerodynamics, which plays a key role in setting the range of an electric vehicle, and the fuel efficiency of an internal combustion engine. The challenge, they realized, was in assembling a dataset of thousands of car designs, each of which is physically accurate in their function and form, without the benefit of physically testing and measuring their performance.
To build a dataset of car designs with physically accurate representations of their aerodynamics, the researchers started with several baseline 3D models that were provided by Audi and BMW in 2014. These models represent three major categories of passenger cars: fastback (sedans with a sloped back end), notchback (sedans or coupes with a slight dip in their rear profile) and estateback (such as station wagons with more blunt, flat backs). The baseline models are thought to bridge the gap between simple designs and more complicated proprietary designs, and have been used by other groups as a starting point for exploring new car designs.
Library of cars
In their new study, the team applied a morphing operation to each of the baseline car models. This operation systematically made a slight change to each of 26 parameters in a given car design, such as its length, underbody features, windshield slope, and wheel tread, which it then labeled as a distinct car design, which was then added to the growing dataset. Meanwhile, the team ran an optimization algorithm to ensure that each new design was indeed distinct, and not a copy of an already-generated design. They then translated each 3D design into different modalities, such that a given design can be represented as a mesh, a point cloud, or a list of dimensions and specs.
The researchers also ran complex, computational fluid dynamics simulations to calculate how air would flow around each generated car design. In the end, this effort produced more than 8,000 distinct, physically accurate 3D car forms, encompassing the most common types of passenger cars on the road today.
To produce this comprehensive dataset, the researchers spent over 3 million CPU hours using the MIT SuperCloud, and generated 39 terabytes of data. (For comparison, it’s estimated that the entire printed collection of the Library of Congress would amount to about 10 terabytes of data.)
The engineers say that researchers can now use the dataset to train a particular AI model. For instance, an AI model could be trained on a part of the dataset to learn car configurations that have certain desirable aerodynamics. Within seconds, the model could then generate a new car design with optimized aerodynamics, based on what it has learned from the dataset’s thousands of physically accurate designs.
The researchers say the dataset could also be used for the inverse goal. For instance, after training an AI model on the dataset, designers could feed the model a specific car design and have it quickly estimate the design’s aerodynamics, which can then be used to compute the car’s potential fuel efficiency or electric range — all without carrying out expensive building and testing of a physical car.
“What this dataset allows you to do is train generative AI models to do things in seconds rather than hours,” Ahmed says. “These models can help lower fuel consumption for internal combustion vehicles and increase the range of electric cars — ultimately paving the way for more sustainable, environmentally friendly vehicles.”
“The dataset is very comprehensive and consists of a diverse set of modalities that are valuable to understand both styling and performance,” says Yanxia Zhang, a senior machine learning research scientist at Toyota Research Institute, who was not involved in the study.
This work was supported, in part, by the German Academic Exchange Service and the Department of Mechanical Engineering at MIT.
In a new dataset that includes more than 8,000 car designs, MIT engineers simulated the aerodynamics for a given car shape, which they represent in various modalities, including “surface fields.”
Login
With over 13 years of experience as a Director of Transportation for a school district, Roche possesses extensive expertise in overseeing logistical operations and ensuring the safety and efficiency of transportation systems. Transitioning into consulting, he has utilized his knowledge to aid school districts in optimizing transportation operations and implementing software solutions. Currently, Roche is committed to collaborating with school districts across the country, assisting them in maximizing the benefits of the software and providing comprehensive consulting services tailored to their transportation requirements.
Carl Allen is an experienced leader in education, transportation, and public policy, currently serving as CEO and founder of 4MATIV Technologies, which he launched in 2018. He previously served as Director of Transportation for Boston Public Schools, Regional Vice President for Transdev in Colorado, and COO/CFO of a charter school network in Minneapolis. Drawing on his training in urban planning and public policy from Harvard’s Kennedy School, and his early experience as a Peace Corps Volunteer teaching high school math in Ghana, Allen supports school districts in tackling complex transportation challenges. He holds degrees in industrial and manufacturing design engineering from Northwestern University and lives in St. Paul, Minnesota with his wife and three children.
